|Publication number||US7093234 B2|
|Application number||US 09/939,005|
|Publication date||Aug 15, 2006|
|Filing date||Aug 24, 2001|
|Priority date||Aug 24, 2001|
|Also published as||US20030041316|
|Publication number||09939005, 939005, US 7093234 B2, US 7093234B2, US-B2-7093234, US7093234 B2, US7093234B2|
|Inventors||Jason D. Hibbeler, Jhy-Chun Wang|
|Original Assignee||International Business Machines Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (24), Non-Patent Citations (12), Referenced by (57), Classifications (10), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
All of the material in this patent application is subject to copyright protection under the copyright laws of the United States and of other countries. As of the first effective filing date of the present application, this material is protected as unpublished material. However, permission to copy this material is hereby granted to the extent that the copyright owner has no objection to the facsimile reproduction by anyone of the patent documentation or patent disclosure, as it appears in the United States Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.
1. Field of the Invention
This invention generally relates to the field of system characterization, and more particularly to CPU (Central Processing Unit) profiling and function call tracing for a target application to enable the identification of program bottlenecks, which cause slow performance.
2. Description of the Related Art
In spite of very fast computer hardware, such as a PowerParallel™ enterprise, and mature operating systems, such as AIX, a given target application's execution performance can be less than optimal. Applying profiling software to the target application on a given enterprise provides clues to answer the question: How can the target application be made to execute faster?
Profiling software is used to identify which portions of the target application are executed most frequently or where most of the time is spent. Profilers are typically used after a basic tool, such as vmstat or iostat commands point out a CPU bottleneck, which is causing slow performance. Tools such as vmstat and iostat report CPU and I/O statistics for the entire system. Using a predetermined benchmark, the profiler analyzes the target application to determine the place or places of the bottlenecks, which result in slow execution. Typically once these bottlenecks of CPU usage or function calls are determined, programming or reprogramming can be employed to reduce the bottleneck or in some cases eliminate it from the target application. These profiling tools, although useful have certain shortcomings. One shortcoming is that profiling tools require the source code of the target application. Many times the source code may not be available to the person running the profiling tests. It is not uncommon for source code to be treated as confidential. The person or entity profiling the software may not be the same entity that wrote the software. Accordingly, a need exists to overcome this problem of requiring the target application source code for profiling.
It is noted that without the source code the profiling study cannot be made. In addition each time a new -pg flag assignment is made the target application must be recompiled and relinked. This recompiling step is time consuming and inhibits the spontaneous “what-if” workflow. It is difficult to just trace part of the target application that is, just 10% of the functions. For example, just 10% of the functions, or 10% of the execution time in a target application. Accordingly, a need exists to overcome these shortcomings and to provide a set of improved profiling tools to run traces with certain diagnostic tools and software probes that allow for optimizing of target applications.
Another shortcoming with the prior art profiling tools is the requirement that any changes to the profiling benchmarks cannot be made once the target application has started. Many times application developers want to examine applications from several perspectives with out being required to re-start the program execution. Accordingly, a need exists to enable changes in the benchmarking tools after the target application has started execution.
Still another shortcoming of the performance and profiling tools available today is the requirement to recompile and/or relink the target application every time the performance and managing tool is used. Typically a -pg flag must be used in the Unix environment. The need to recompile and/or relink the source code with special debugging flags many times restricts the user from making timely or spontaneous changes to the application. Each time the -pg flag is changed the application must be recompiled and relinked. Accordingly, a need exists to provide a solution to overcome this shortcoming as well.
Yet another shortcoming with the prior art performance profiling tools is how the results of a function trace are reported. Today, each function in a file compiled with -pg will have a corresponding entry in the gmon.out file. Since the choice of what to profile can only be done at the file level, this could potentially leas to a lot of unwanted data.
The trace output file in format of a gmon.out file does have a set of tools that are used to further identify and understand the location of the bottlenecks. It is desirable for any new and improved trace characterization technique to output the results in the gmon.out file format, which is familiar to the user and allows for continued usage of the characterization tools.
Accordingly, a need exists for a trace characterization technique that will not only eliminate all of the shortcomings listed above but also maintain compatibility with existing output and analysis tool formats.
Briefly, according to the present invention, disclosed is a method, a system and computer readable medium for characterizing a target application using DPCL (Dynamic Probe Class Library) instrumentation, without the need for the source code, or any recompiling or relinking. The instrumentation consists of the selection of suspected hot spots or bottlenecks in the target application and dynamically patching the code to insert calls to the monitor( ) and mcount( ) functions or their equivalents, based on the operating system being used. The characterization can be applied while the target application is running. The characterization output is presented in a gmon.out format.
The method for profiling a target application running on an informational processor begins with applying DPCL (Dynamic Probe Class Library) instrumentation. The DPCL instrumentation applied includes selecting at least one function in the target application to be traced. The method of profiling a target application running on an informational processor begins with applying DPCL instrumentation. The DPCL instrumentation selects at least one function in the application and dynamically patches in calls to the appropriate performance-gathering interfaces. Next, the application is started (if it is not already running), and the results are then written out in gmon.out format for the selected functions.
The subject matter, which is regarded as the invention, is particularly pointed out and distinctly claimed in the claims at the conclusion of the specification. The foregoing and other objects, features, and advantages of the invention will be apparent from the following detailed description taken in conjunction with the accompanying drawings.
It is important to note that these embodiments are only examples of the many advantageous uses of the innovative teachings herein. In general, statements made in the specification of the present application do not necessarily limit any of the various claimed inventions. Moreover, some statements may apply to some inventive features but not to others. In general, unless otherwise indicated, singular elements may be in the plural and visa versa with no loss of generality.
In the drawing like numerals refer to like parts through several views.
The present invention as would be known to one of ordinary skill in the art could be produced in hardware or software, or in a combination of hardware and software. However in one embodiment the invention is implemented in software. The system, or method, according to the inventive principles as disclosed in connection with the preferred embodiment, may be produced in a single computer system having separate elements or means for performing the individual functions or steps described or claimed or one or more elements or means combining the performance of any of the functions or steps disclosed or claimed, or may be arranged in a distributed computer system, interconnected by any suitable means as would be known by one of ordinary skill in the art.
According to the inventive principles as disclosed in connection with the preferred embodiment, the invention and the inventive principles are not limited to any particular kind of computer system but may be used with any general purpose computer, as would be known to one of ordinary skill in the art, arranged to perform the functions described and the method steps described. The operations of such a computer, as described above, may be according to a computer program contained on a medium for use in the operation or control of the computer, as would be known to one of ordinary skill in the art. The computer medium, which may be used to hold or contain the computer program product, may be a fixture of the computer such as an embedded memory or may be on a transportable medium such as a disk, as would be known to one of ordinary skill in the art.
The invention is not limited to any particular computer program or logic or language, or instruction but may be practiced with any such suitable program, logic or language, or instructions as would be known to one of ordinary skill in the art. Without limiting the principles of the disclosed invention any such computing system can include, inter alia, at least a computer readable medium allowing a computer to read data, instructions, messages or message packets, and other computer readable information from the computer readable medium. The computer readable medium may include non-volatile memory, such as ROM, Flash memory, floppy disk, Disk drive memory, CD-ROM, and other permanent storage. Additionally, a computer readable medium may include, for example, volatile storage such as RAM, buffers, and cache memory.
Exemplary Hardware for a Multi-Computer System
In one embodiment, the techniques of the present invention are used in distributed computing environments in order to provide multi-computer applications. These applications are used in very demanding applications such as finance, computational chemistry, bioinformatics, weather prediction and even military types of applications. These applications are very complex and are being used in a multi-computer environment. In order to reduce the processing time and improve the ability to make even finer characterization runs, every effort is made to assure that the application has been optimized and that any bottlenecks are eliminated. One example of the hardware that runs these types of applications is the IBM RISC System/6000 Scalable PowerParallel™ systems, also known as the SP system.
An “N” way Multiprocessing Enterprise
In one example, distributed computing environment 200 includes N nodes 202 with one or more processors 206. In one instance, each processing node is a RISC/6000 computer running AIX, the IBM version of the UNIX operating system. The processing nodes do not have to be RISC/6000 computers running the AIX operating system. Some or all of the processing nodes 202 can include different types of computers and/or different Unix based operating systems 308. All of these variations are considered a part of the claimed invention.
Exemplary Software for a Multi-Computer System
Flow Diagram of a Trace Characterization
As described in the glossary, the term DPCL (IBM's Dynamic Probe Class Library) is just one mechanism for inserting dynamic instrumentation (i.e., changing the program selectively while it is running). It should be understood of to those of average skill in the art, that the present invention can be implemented using a different framework for dynamic instrumentation other than DPCL within the true scope and spirit of the present invention.
The target application may then be started or may have already been running. Unlike the prior art, the trace application is dynamic and can be applied to the target application at any time and without re-compiling or relinking. Once the user decides that sufficient data has been gathered, the results are examined at step 410. This examination can be “real-time”: the results can be viewed as the data is being produced. This output file can be analyzed using compatible profiling tools according to the prior art. If the study is determined to be completed at step 412 the flow exits at step 414. If there is a need to modify the DPCL tool, according to step 416, for additional analysis this is accomplished at step 418 and the target application is executed again starting at step 408. If the DPCL tool is not modified at step 416 the target application may be re-started for additional trace information based on new benchmarks or a different platform set of parameters. This is repeated on the target application, until the analysis is finished and the flow exits at 414.
For a given target application, the function call-count and CPU usage information is collected. Using the DPCL class library, the user can construct a separate DPCL tool that can examine the target application in a non-invasive way. In other words, the user does not need to statically instrument the target application by recompiling and relinking it with the -pg compiler/link flag. Instead, the application is started directly or the DPCL tool would start the target application and then the DPCL analysis tool connects to the application. This is in much the same way that a debugger connects to a target application. This collects the information dynamically as the program runs, saves the information in standard gmon.out format, and then disconnects from the application.
The DPCL class library itself encapsulates and hides the low-level mechanisms of connecting to the target application and examining the application. It is a straightforward matter to take advantage of the flexibility that the DPCL affords to combine DPCL with the standard pieces of process profiling in UNIX to create a dynamic non-invasive profiling tool.
High Level CPU Profiling Control Flow Diagram
Turning now to
When execution hits a CPU profiling “on” point and the CPU profiling is currently off, the CPU profiling is turned on and a message is sent back to the DCPL tool to indicate CPU profiling is turned on, at step 440. When execution hits a CPU profiling “off” point and the CPU profiling is currently on, the CPU profiling is tuned off and a message is sent back to the DPCL tool to indicate that CPU profiling is turned off at step 442.
Once the target application has been running for sufficient time a DPCL one-time probe can be used to retrieve a intermediate report at step 444. Note the profiling can be stopped at step 446. Now at step D 448 the operator can decide to loop back to step B 428 and either re-enter a profiling run, or at point C step 434 re-select functions for profiling and functional call count tracing.
Finally the profiling is completed at step 446. If needed at point D 448 the target application can be re-engaged at node a 422. Alternatively, different DPCL probe points for intermediate reports can be specified at point C 434. If no new or different profiling is desired at points B or C the present invention disconnects from the application at step 450. At point E 452 an entirely different target application can be selected and connected to at point A 442; if not the flow exits at step 454.
The following discussion is provided for those skilled in the art to be able to use the present invention.
Before a profiling study can begin to locate hot spots in a target application, the target application must be fully functional and have realistic data values to be profiled with. A key command used in the profiling is the prof command. The prof command displays a profile of CPU usage for each external symbol or routine of a target application. In detail, it displays the following:
The prof command interprets the profile data collected by the monitor( ) subroutine for the object file (a.out by default), reads the symbol table in the object file, and correlates it with the profile file (mon.out by default) generated by the monitor( ) subroutine. A usage report is sent to the terminal or can be redirected to a file.
To use the prof command, the -p option is used to compile a source program in C, FORTRAN, PASCAL, or COBOL. This inserts a special profiling startup function into the object file that calls the monitor( ) subroutine to track function calls. When the program is executed, the monitor( ) subroutine creates a mon.out file to track execution time. Therefore, only programs that explicitly exit or return from the main program cause the mon.out file to be produced. Also, the -p flag causes the compiler to insert a call to the mcount( ) subroutine or its equivalent (depending on the operating system being used) into the object code generated for each recompiled function of the program. While the program runs, each time a parent calls a child function, the child calls the mcount( ) subroutine to increment a distinct counter for that parent-child pair. This counts the number of calls to a function.
By default, the displayed report is sorted by decreasing percentage of CPU time. This is the same as when specifying the -t option.
The -c option sorts by decreasing number of calls and the -n option sorts alphabetically by symbol name.
If the -s option is used, a summary file mon.sum is produced. This is useful when more than one profile file is specified with the -m option (the -m option specifies files containing monitor data).
The -z option includes all symbols, even if there are zero calls and time associated.
Other options are available and explained in the prof command in the AIX Commands Reference.
Turning now to
Line 502 of table 500 contains the headings, and describing them from left to right:
Lines 516 list the example output with all of the subroutine calls. Given this list, the question is: Are the subroutines using an appropriate amount of execution time?
It is also noted that the calls to the different subroutines are summarized here. The previous art would list each and every call. This would result in a very large output file with no real added value.
In this example, many calls to the mod8( ) line 518 and mod9( ) line 520 routines are made. With this as a starting point, the source code would be examined to see why they are used so much. Another starting point could be to investigate why a routine requires so much time. With these starting points one skilled in the art can tune and optimize the target application using the subject invention.
The Gprof Command
The gprof command produces an execution profile of C, PASCAL, FORTRAN, or COBOL programs. The statistics of called subroutines are included in the profile of the calling program. The gprof command is useful in identifying how a program consumes CPU resources. It is roughly a superset of the prof command, giving additional information and providing more visibility to active sections of code.
The Gprof Implementation
The source code must be compiled with the -pg option. This action links in versions of library routines compiled for profiling and reads the symbol table in the named object file (a.out by default), correlating it with the call graph profile file (gmon.out by default). This means that the compiler inserts a call to the mcount( ) function into the object code generated for each recompiled function of the target application. The mcount( ) function counts each time a parent calls a child function. Also, the monitor( ) function is enabled to estimate the time spent in each routine.
The gprof command generates two useful reports:
Each report section begins with an explanatory part describing the output columns. These pages can be suppressed by using the -b option.
Use -s for summaries and -z to display routines with zero usage.
Where the program is executed, statistics are collected in the gmon.out file. These statistics include the following:
When the gprof command is issued, it reads the a.out and gmon.out files to generate the two reports. The call-graph profile is generated first, followed by the flat profile. It is best to redirect the gprof output to a file, because browsing the flat profile first may answer most questions about the target application.
Cwhet Benchmark Program Output File
Turning now to
The call-graph profile is the first part of the cwhet.gprof file and looks similar to
Flat Profile of Cwhet.Gprof Sample
Turning now to
The flat profile is much less complex than the call-graph profile of
Normally, the top functions on the list are candidates for optimization. However, care should be taken to also consider how many calls are made to the function. Sometimes it can be easier to make slight improvements to a frequently called function than to make extensive changes to a piece of code that is called once.
Cross Reference Index
Turning now to
Glossary of Terms Used in this Disclosure
Although a specific embodiment of the invention has been disclosed, it will be understood by those having skill in the art that changes can be made to this specific embodiment without departing from the spirit and scope of the invention. The scope of the invention is not to be restricted, therefore, to the specific embodiment, and it is intended that the appended claims cover any and all such applications, modifications, and embodiments within the scope of the present invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5121501||Dec 27, 1989||Jun 9, 1992||International Business Machines Corporation||First processor inserting hooks into software and sending unique identifications to output bus and second processor associating data frames and time with these unique identifications|
|US5347649||Mar 6, 1990||Sep 13, 1994||International Business Machines Corp.||System for dynamically generating, correlating and reading multiprocessing trace data in a shared memory|
|US5355487||Jul 23, 1993||Oct 11, 1994||International Business Machines Corporation||Non-invasive trace-driven system and method for computer system profiling|
|US5446876||Apr 15, 1994||Aug 29, 1995||International Business Machines Corporation||Hardware mechanism for instruction/data address tracing|
|US5594864||May 30, 1995||Jan 14, 1997||Sun Microsystems, Inc.||Method and apparatus for unobtrusively monitoring processor states and characterizing bottlenecks in a pipelined processor executing grouped instructions|
|US5664095||Dec 28, 1994||Sep 2, 1997||Intel Corporation||Dynamic scaling of CPU cycle consumption in a computer system|
|US5761091||Dec 10, 1996||Jun 2, 1998||Bgs Systems, Inc.||Method and system for reducing the errors in the measurements of resource usage in computer system processes and analyzing process data with subsystem data|
|US5797115||Aug 24, 1995||Aug 18, 1998||Fuller; Billy||Measuring direct and indirect usage of a central processing unit|
|US5850562 *||Jun 27, 1994||Dec 15, 1998||International Business Machines Corporation||Personal computer apparatus and method for monitoring memory locations states for facilitating debugging of post and BIOS code|
|US5884080 *||Nov 26, 1996||Mar 16, 1999||International Business Machines Corporation||System and method for instruction burst performance profiling for single-processor and multi-processor systems|
|US5890014 *||Aug 5, 1996||Mar 30, 1999||Micronet Technology, Inc.||System for transparently identifying and matching an input/output profile to optimal input/output device parameters|
|US5907704 *||Oct 2, 1996||May 25, 1999||Quark, Inc.||Hierarchical encapsulation of instantiated objects in a multimedia authoring system including internet accessible objects|
|US5950000 *||Dec 6, 1996||Sep 7, 1999||Sun Microsystems, Inc.||Integrated of a third party software tool|
|US5964893||Aug 30, 1995||Oct 12, 1999||Motorola, Inc.||Data processing system for performing a trace function and method therefor|
|US6086618||Jan 26, 1998||Jul 11, 2000||Microsoft Corporation||Method and computer program product for estimating total resource usage requirements of a server application in a hypothetical user configuration|
|US6189142 *||Sep 16, 1998||Feb 13, 2001||International Business Machines Corporation||Visual program runtime performance analysis|
|US6353923 *||Jan 30, 1998||Mar 5, 2002||Microsoft Corporation||Active debugging environment for debugging mixed-language scripting code|
|US6412106 *||Jun 16, 1999||Jun 25, 2002||Intervoice Limited Partnership||Graphical system and method for debugging computer programs|
|US6708169 *||Jul 10, 2000||Mar 16, 2004||International Business Machines Corporation||Apparatus and method for generating a merged symbol file for verifying symbolic data|
|US6708327 *||Oct 14, 1999||Mar 16, 2004||Techonline, Inc.||System for accessing and testing evaluation modules via a global computer network|
|US6728949 *||Dec 14, 1999||Apr 27, 2004||International Business Machines Corporation||Method and system for periodic trace sampling using a mask to qualify trace data|
|US6848097 *||Sep 18, 2002||Jan 25, 2005||Cray Inc.||Debugging techniques in a multithreaded environment|
|US6981248 *||May 2, 2002||Dec 27, 2005||International Business Machines Corporation||Conditional breakpoint encountered indication|
|JPH06342386A||Title not available|
|1||*||Derose et al., The Dynamic Probe Class Library-An Infrastructure for developing instrumentation . . . , IEEE, pp. (all).|
|2||*||Dmitriev, "Profiling java applications using code hotswapping and dyanmic call graph revelation" ACM WOSP,, 139-150, 2004.|
|3||IBM Research Disclosure No. 42898, "Automated Inclusion of Program Tracing Statements," Dec. 1999.|
|4||IBM Research Disclosure No. 431, "Dynamic Monitoring of System Parameters," Mar. 2000.|
|5||IBM Technical Disclosure Bulletin, "Process for Real-Time, Trace-Driven Performance Monitors," vol. 34, No. 5, Oct. 1991.|
|6||IBM Technical Disclosure Bulletin, "Processor Architecture for Measurement and Monitoring Functions," vol. 32, No. 9A, Feb. 1990.|
|7||IBM Technical Disclosure Bulletin, Programmable Performance Monitor Apparatus for a Multi-Chip Super Scalar Processor, vol. 37, No. 04B, Apr. 1994.|
|8||*||Ingles, Dynamic Probe Class Library-A Parallel Tools Consortium Project, Aug. 30, 2000, pp. 1-4.|
|9||*||Joshi et al, "Targeted path profiling: lower overhead path profiling for staged dynamic optimization systems", IEEE CGO, pp. 1-12, 2004.|
|10||*||Lawrence Livermore National Laboratory, Performance Analysis Tools, Jun. 27, 2001, sections 4 and 6.|
|11||*||Odom et al, "Using dynamic tracing sampling to measure long running programs", IEEE SC, pp. 1-10, 2005.|
|12||*||Wang et al, "Using compressed bytecode traces for slicing java programs", IEEE ICSE, pp. 1-10, 2004.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7367025||Dec 30, 2003||Apr 29, 2008||Sap Ag||Byte code modification for testing, debugging and/or monitoring of virtual machine based software|
|US7426723||Dec 30, 2003||Sep 16, 2008||Sap Ag||Classfile conversion into an organization of objects, and modification thereof, to effect bytecode modification|
|US7506315||Dec 30, 2003||Mar 17, 2009||Sap Ag||System and method combining application tracing and distributed statistical records|
|US7526550||Mar 26, 2004||Apr 28, 2009||Sap Ag||Unified logging service with a log viewer|
|US7577731||Dec 30, 2003||Aug 18, 2009||Sap Ag||System and method for monitoring program code|
|US7640539||Apr 12, 2005||Dec 29, 2009||International Business Machines Corporation||Instruction profiling using multiple metrics|
|US7644395||Dec 30, 2003||Jan 5, 2010||Sap Ag||System and method employing bytecode modification techniques for tracing services within an application server|
|US7657875 *||Apr 12, 2005||Feb 2, 2010||International Business Machines Corporation||System and method for collecting a plurality of metrics in a single profiling run of computer code|
|US7689558||Sep 9, 2005||Mar 30, 2010||Sap Ag||Application monitoring using profile points|
|US7707557 *||Dec 30, 2003||Apr 27, 2010||Sap Ag||Execution of modified byte code for debugging, testing and/or monitoring of object oriented software|
|US7714864||Nov 16, 2006||May 11, 2010||Microsoft Corporation||Visual resource profiler for graphical applications|
|US7721266||Mar 26, 2004||May 18, 2010||Sap Ag||Unified logging service with a logging formatter|
|US7725572||Dec 30, 2003||May 25, 2010||Sap Ag||Notification architecture and method employed within a clustered node configuration|
|US7743367||Dec 30, 2003||Jun 22, 2010||Sap Ag||Registration method for supporting bytecode modification|
|US7756968||Dec 30, 2003||Jul 13, 2010||Sap Ag||Method and system for employing a hierarchical monitor tree for monitoring system resources in a data processing environment|
|US7810075||Apr 29, 2005||Oct 5, 2010||Sap Ag||Common trace files|
|US7822826||Dec 30, 2003||Oct 26, 2010||Sap Ag||Deployment of a web service|
|US7836438 *||Dec 30, 2003||Nov 16, 2010||Sap Ag||Modified classfile registration with a dispatch unit that is responsible for dispatching invocations during runtime execution of modified bytecode|
|US7895580||Dec 30, 2003||Feb 22, 2011||Sap Ag||Application tracing service employing different levels of precision for modifying bytecode|
|US7941521||Dec 30, 2003||May 10, 2011||Sap Ag||Multi-service management architecture employed within a clustered node configuration|
|US7954095 *||Nov 1, 2006||May 31, 2011||International Business Machines Corporation||Analysis and selection of optimal function implementations in massively parallel computer|
|US8024711 *||Aug 10, 2005||Sep 20, 2011||Microsoft Corporation||Software analysis tool|
|US8176480 *||Feb 27, 2006||May 8, 2012||Symantec Operating Corporation||Adaptive instrumentation through dynamic recompilation|
|US8214819 *||Jul 9, 2004||Jul 3, 2012||Hewlett-Packard Development Company, L.P.||Determining call counts in a program|
|US8286135 *||Oct 17, 2007||Oct 9, 2012||Cray Inc.||Performance visualization including hierarchical display of performance data|
|US8566796 *||Apr 4, 2008||Oct 22, 2013||Sas Institute Inc.||Systems and methods for interactions with software probes|
|US8639982||Jun 20, 2011||Jan 28, 2014||International Business Machines Corporation||Dynamic computer process probe|
|US9058421 *||Jun 16, 2009||Jun 16, 2015||Freescale Semiconductor, Inc.||Trace correlation for profiling subroutines|
|US9058424 *||Oct 29, 2012||Jun 16, 2015||The Mathworks, Inc.||Automatic unit test generation and execution|
|US9098625 *||Feb 29, 2008||Aug 4, 2015||International Business Machines Corporation||Viral trace|
|US9298589||Feb 23, 2015||Mar 29, 2016||Microsoft Technology Licensing, Llc||User interaction analysis of tracer data for configuring an application tracer|
|US9389992 *||Aug 7, 2015||Jul 12, 2016||Microsoft Technology Licensing, Llc||Multiple tracer configurations applied on a function-by-function level|
|US9417993 *||Aug 7, 2015||Aug 16, 2016||Microsoft Technology Licensing, Llc||Real time analysis of tracer summaries to change tracer behavior|
|US9552202||Dec 8, 2011||Jan 24, 2017||International Business Machines Corporation||Automated and heuristically managed solution to quantify CPU and path length cost of instructions added, changed or removed by a service team|
|US9575874||Mar 9, 2015||Feb 21, 2017||Microsoft Technology Licensing, Llc||Error list and bug report analysis for configuring an application tracer|
|US9658936||Dec 24, 2014||May 23, 2017||Microsoft Technology Licensing, Llc||Optimization analysis using similar frequencies|
|US9658943||Mar 23, 2015||May 23, 2017||Microsoft Technology Licensing, Llc||Interactive graph for navigating application code|
|US9665474||Mar 29, 2013||May 30, 2017||Microsoft Technology Licensing, Llc||Relationships derived from trace data|
|US20040194066 *||Dec 30, 2003||Sep 30, 2004||Frey Gregor K||System and method for monitoring program code|
|US20050216585 *||Mar 26, 2004||Sep 29, 2005||Tsvetelina Todorova||Monitor viewer for an enterprise network monitoring system|
|US20050223282 *||Mar 26, 2004||Oct 6, 2005||Frey Gregor K||Unified logging service with a log viewer|
|US20050223283 *||Mar 26, 2004||Oct 6, 2005||Frey Gregor K||Unified logging service with a logging formatter|
|US20060020918 *||Jul 9, 2004||Jan 26, 2006||David Mosberger||Determining call counts in a program|
|US20060129990 *||Dec 13, 2004||Jun 15, 2006||International Business Machines Corporation||System, method, and article of manufacture for determining execution counts associated with a software program of an embedded system|
|US20060174226 *||Feb 3, 2005||Aug 3, 2006||Sytex, Inc.||Methods, Test Systems And Computer-Readable Medium For Dynamically Modifying Flow Of Executable Code|
|US20060230390 *||Apr 12, 2005||Oct 12, 2006||International Business Machines Corporation||Instruction profiling using multiple metrics|
|US20060230391 *||Apr 12, 2005||Oct 12, 2006||International Business Machines Corporation||System and method for collecting a plurality of metrics in a single profiling run of computer code|
|US20060248177 *||Apr 29, 2005||Nov 2, 2006||Sap Aktiengesellschaft||Common trace files|
|US20070038974 *||Aug 10, 2005||Feb 15, 2007||Microsoft Corporation||Software analysis tool|
|US20070074170 *||Sep 9, 2005||Mar 29, 2007||Rossmann Paul A||Application monitoring using profile points|
|US20080092121 *||Oct 17, 2007||Apr 17, 2008||Cray Inc.||Performance visualization including hierarchical display of performance data|
|US20080101232 *||Nov 1, 2006||May 1, 2008||Charles Jens Archer||Analysis and selection of optimal function implementations in massively parallel computer|
|US20090222646 *||Feb 28, 2008||Sep 3, 2009||International Business Machines Corporation||Method and apparatus for detecting processor behavior using instruction trace data|
|US20090222796 *||Feb 29, 2008||Sep 3, 2009||International Business Machines Corporation||Viral trace|
|US20090254821 *||Apr 4, 2008||Oct 8, 2009||Sas Institute Inc.||Systems And Methods For Interactions With Software Probes|
|US20100275185 *||Apr 24, 2009||Oct 28, 2010||Felix Burton||System and Method for High Performance Coverage Analysis|
|US20100318972 *||Jun 16, 2009||Dec 16, 2010||Freescale Semiconductor, Inc.||Trace Correlation for Profiling Subroutines|
|U.S. Classification||717/124, 717/127, 717/128|
|International Classification||G06F9/44, G06F9/45|
|Cooperative Classification||G06F11/3466, G06F11/3423, G06F2201/865|
|European Classification||G06F11/34T, G06F11/34C4A|
|Aug 24, 2001||AS||Assignment|
Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:HIBBELER, JASON D.;WANG, JHY-CHUN;REEL/FRAME:012130/0356;SIGNING DATES FROM 20010820 TO 20010823
|Mar 22, 2010||REMI||Maintenance fee reminder mailed|
|Aug 15, 2010||LAPS||Lapse for failure to pay maintenance fees|
|Oct 5, 2010||FP||Expired due to failure to pay maintenance fee|
Effective date: 20100815